European Journal of Scientific Research ISSN 1450-216X Vol.49 No.4 (2011), pp. 590-600 EuroJournals Publishing, Inc.

. 2011 http://www.eurojournals.com/ejsr.htm

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies
Seddik Bri Microwave and Materials Group, ESTM, B. P 3103, Mekns Morocco Tel: (00)(212) 5 35 46 70 86, Fax: (00) (212) 5 35 46 70 83 E-mail: briseddik@gmail.com Samira Kassimi Laboratory Systems and Telecommunications Engineering Decision Faculty des Sciences de Kenitra - Morocco E-mail: kassimisam@yahoo.fr Mohamed Habibi Laboratory Systems and Telecommunications Engineering Decision Faculty des Sciences de Kenitra - Morocco E-mail: kassimisam@yahoo.fr Ahmed Mamouni Institute of Electronics, Microelectronics and Nanotechnology, UMR CNRS 8520 IEMN Cit Scientifique, Avenue Poincar - B.P 60069 59652, Villeneuve dAscq, Lille- France. Abstract The papers purpose estimate the specific absorption rate (SAR) distribution in a human head exposed to electromagnetic field emitted from a handheld cellular phone operating in the 900 MHz and 1800 MHz rang in a partially closed environment. The HFSS software called high frequency structure simulator (HFSS) and the finite-difference timedomain (FDTD) computations were used to evaluate the SAR in realistic anatomically based model of the human head for different antenna-head distances.

Keywords: SAR, GSM, HFSS, Human head, FDTD, Dipole antenna

1. Introduction
When electromagnetic waves hit a biological tissue, the wave is absorbed, refracted and diffracted. The physical modification of matter after interaction can have a biological effect. Thus, the microwave heating of tissue, constituted in great part of water, causes a damage of the tissue if the temperature elevation is too intense or prolonged. The interaction depends on the dielectric matter properties (conductivity and permittivity) and the rate between the wave-length and tissue dimensions [1-2]. In practice, characteristics of the heat effect are evaluated by the absorbed electric power per units mass in the tissue, i.e., the specific absorption rate (SAR) (in watts per kilogram). The local 1-or 10-g average SAR has been used for the primary dosimeter parameter of the electromagnetic waves

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies

591

exposure [3]. These SARs are generally estimated from numerical simulation [4-7] and experimental evaluation based on the E-field and thermal measurement [8-10]. In this paper, we present the simulation results of the absorption of RF energy in the biological tissue in the case of the cell phones. The Ansoft HFSS method with boundary conditions and assigned excitations by an antenna dipole, an antenna monopole and patch antenna are used as the cell phone. The human head is simulated by HFSS at GSM frequencies 900 MHz and 1800 MHz. In order to evaluate the ability of the HFSS code, we use the FDTD computations and the measurement results of the SAR distribution obtained from the spherical glass bowl filled with different liquids and either a wired dipole antenna, monopole antenna or patch antenna which is placed below the bowl for inducing EM fields in the liquid [11].

2. Used Method and Modelling

In the first time we consider a sphere consisting entirely of material with the electric properties of brain-equivalent tissue with parameters permittivity r and conductivity (figure1).
Figure 1: Homogeneous sphere with cylindrical boundary conditions.

The sphere is irradiated by a dipole antenna of length 2L (L depends on the frequency) with a gap width of 1 mm located at different distance from the sphere. The conducting medium is surround by a glass shell with at thickness of 5 mm and dielectric constant r =4,6 representing the recipient of the brain-equivalent liquid in testing situations. The radiated power of the dipole in presence of the sphere is 1W (continuous-wave mode). In the second time, we consider a sphere consisting entirely of material with the electric properties of skin equivalent tissue. In the third time, the model comprised three layers are considered. The spherical model had a uniform content at its core (representing the human brain) and the core was surround by two spherical shells representing the skull (bone) and the muscle with their respective electromagnetic properties (figure 2). In the last time, the model comprised four layers are considered. In this simulation, the core was surrounding by three spherical shells representing the skin, the skull (bone) and the muscle with their respective electromagnetic properties. The properties of the materials used in the simulations are presented in table 1 [12].

592

Seddik Bri, Samira Kassimi, Mohamed Habibi and Ahmed Mamouni

Figure 2: Layered sphere.

Table 1:

The properties of the materials used in the simulations.

Material 900 MHZ 39,5 55,03 12,5 56,8 2,7 900 MHZ (S/m) 0,7 0,943 0,17 1,1 0,0016 1800 MHZ 38,2 53,55 12 51,8 2,7 1800 MHZ (S/m) 0,9 1,34 0,29 1,5 0,003 Mass Density (Kg/m3) 1080 1040 1180 1050 -

Skin Muscle Bone Brain (cortical) Dielectric phone cover

The distribution of the local SAR values can be calculated directly from the total electric field distribution E(V/m), which results from the HFSS code. The rate at which the energy is absorbed in tissue par unit mass is called the specific absorption rate (SAR) (W/kg): 2 d dW .|E | SAR = ( )= dt dm 2. (1) is the conductivity in mhos/m, and the density of the tissue in kg/m3 (this is usually given in units of g/cm3, e.g., for water, =1g/cm3). SAR is calculated as a function of position from the estimates of local fields and tissues properties. An integral of SAR over a volume of tissue containing a given mass gives the absorbed power. This is typically expressed in units of mW/g, or mW/cm3 (for a given tissue density) averaged over 1g of tissue [13]. The SAR value averaged over 10g was subsequently calculated considering the contribution of the smaller cube and the contribution of the cubical shell around it:

(SAR)i mi + (SAR) j m j
SAR( 10g ) =
V1 V2 V1

mi +
V1

mj
V2 V1

(2) 10 V1 m j = Pj V V2 - V1 . Index i refers to the lattice cells inside the inner Where mi= x V and cube and index j to those around it.

3. Finite Difference Time Domain (FDTD) Computations

The spherical bowl and the dipole were modelled in a cubical FDTD grid with grid step equal to 2.5 mm, as shown in figure 3. This grid step was chosen as suitable for computing distances 5, 25 and 50 mm between the bowl and the antenna but also giving moderate modelling errors for the dimensions of both structures. Obviously, in order to have a symmetrical antenna, the length of the antenna model is always an odd number of cells and therefore the diameter of the bowl also has to be an odd number of cells if the antenna is to be placed in a true centred position below the bowl. This requires though that

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies

593

the antenna is modelled as a bar of cells rather than by a thin filament of FDTD components if the models are to be symmetrical also in the plane perpendicular to the antenna axis. However, when modelling the case with a asymmetrically positioned antenna the tip of the dipole is not possible to placed directly under the outer south pole but it will be a half grid step offset from this position. The FDTD components in the glass-liquid boundary, i.e. on the inside of the bowl, were computed with the material parameters set equal to those for the liquid since the pyrex glass has a zero conductivity. In the 2,5 mm grid, the bowl has an outer diameter of 89 cells, i.e. 222,5 mm, and an inner diameter of 85 cells, i.e. 212.5 mm. The antenna is represented by two bars each 33 cells long with a one by one cell cross section giving an overall length, including the voltage source gap, of 67 cells or 167,5 mm (figure 3).
Figure 3: The dimensions of the FDTD models. The separation h between the bowl and the dipole was 2, 10 and 20 cells corresponding to distances 5, 25 and 50 mm.

The bowl and the half wave dipole were placed in the FDTD grid with a minimum distance to the boundary of l/3 giving a total computational volume of 165x165x165 cells for the computations with the dipole in a centred position and 165x190x165 cells for the case when it was placed asymmetrically.

4. Results
The spherical phantom is a glass bowl with an opening at it northern hemisphere. The opening allows a probe to move within the liquid. It has been determined that the opening does not alter the fields in the southern hemisphere: Inner radius =106,55mm, thickness = 50,5mm, opening =d2=170 mm and r =4,6. At 900 MHz, the first model (homogeneous sphere: figure 1), the conducting medium has a r of 41.5 and a = 0,97 S/m. The Inner radius=106,55mm, thickness =5 0,5mm, opening =d2=170mm. The following liquid dielectric properties were measured are: r=43,32 5%, =0.86210%,=1,33 kg/m3 (For the SAR evaluations, the density will be assumed to be 1 kg/m3), Level = d1 = 13.4 cm (figure 1). The gap between dipoles = 1mm and wire diameter = 3,6mm. The length L of on arm of the antenna is 74 mm. At 1800 MHz, the conducting medium has an r of 40,5 and a =1,40 S/m. The length L of on arm of the antenna is 35,5 mm. The same model of head has been considered, but this time it is excited by a monopole to go up on a metallic case. The length L of arm of the antenna at 900 MHz is 79,5 mm and a diameter is 3,6mm. At 1800 MHz, The length L of arm of the antenna is 39.5 mm and the same diameter.

594

Seddik Bri, Samira Kassimi, Mohamed Habibi and Ahmed Mamouni

With the same model also, it is excited by a rectangular patch antenna. The measurements of the antenna at 900 MHz are: the length is 97,87 mm and the width is 75,92. The patch is fed by a microstrip line, the length of line is 45,01 mm and the width is 2,86 mm. At 1800 MHz, The measurements of the antenna are: the length is 48,93 mm and the width is 37,77. The patch is fed by a microstrip line, the length of line is 22 mm and the width is 2,86 mm. The figure 4 gives the local and average SAR variation in the first model (homogenous sphere: Brain). The figure 5 gives the local and average SAR variation in the skin. Figure 6 gives the distribution of SAR in the layered sphere: brain (60mm), bone (10mm), and skin (10mm). The figure 7 gives the local and average SAR variation in the layered sphere: skin (5mm), muscle (5mm), bone (5mm) and brain (95mm).The HFSS evaluation method of the local and average SAR is based on distribution on the Electric field. These SARs are calculated when depth d varied and normalized to maximal value. Notice that the SAR is higher in homogeneous head phantoms than in heterogeneous.
Figure 4: Local and average SAR variation, at different antenna-head distances, for homogenous sphereBrain:
(a) Dipole at 900 MHz (c) Monopole at 900 MHz
20
Local_SAR: 5mm Average_SAR: 5mm Local_SAR:10mm

(b) Dipole at 1800 MHz (d) Monopole at 1800 MHz

60
Local_SAR: 5mm Average_SAR: 5mm Local_SAR: 10mm Average_SAR: 10mm Local_SAR: 20mm Average_SAR: 20mm

50

15

Average_SAR:10mm Local_SAR: 20mm Average_SAR: 20mm

40

10

(a)

30

(b)

20

5
10

0 0 0,2 0,4 0,6 Normalized distance 0,8 1

0 0 0,2 0,4 0,6 0,8 1 Normalized distance

16 14

120

(c)
12 10 8 6 4

SAR in W/Kg

SAR in W/Kg

Local_SAR_5mm Average_SAR_5mm Local_SAR_10mm Average_SAR_10mm Local_SAR_15mm Average_SAR_15mm Local_SAR_20mm Average_SAR_20mm

100

80

(d)

Local_SAR_5mm Average_SAR_5mm Local_SAR_10mm Average_SAR_10mm Local_SAR_15mm Average_SAR_15mm Local_SAR_20mm Average_SAR_20mm

60

40

20
2 0 0,0 0,2 0,4 0,6 0,8 1,0 Normalized distance

0 0,0 0,2 0,4 0,6 0,8 1,0

Normalized distance

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies

595

Figure 5: Local and average SAR variation, at different antenna-head distances, for homogenous sphere Skin:
(a) Dipole at 900 MHz (c) Monopole at 900 MHz (e) Patch at 900 MHz
25

(b) Dipole at 1800 MHz (d) Monopole at 1800 MHz (f) Patch at 1800 MHz
70

20

(a)

Local_SAR:5m m Average_SAR:5m m Local_SAR:10m m Average_SAR:10mm Local_SAR:20m m Average_SAR:20mm

SAR in W/Kg

60

(b)

50

SAR in W/Kg

15
40

Local_SAR:5mm Average_SAR:5mm Local_SAR:10mm Average_SAR:10mm Local_SAR:15mm Average_SAR:15mm Local_SAR:20mm Average_SAR:20mm

10

30

20

5
10

0 0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

0,0

0,2

0,4

0,6

0,8

1,0

Norm alized distance

14
35

Normalised Distance

12

10

(c)

SAR in W/Kg

SAR in W/Kg

Local_SAR_5mm Average_SAR_5mm Local_SAR_10mm Average_SAR_10mm Local_SAR_15mm Average_SAR_15mm Local_SAR_20mm Average_SAR_20mm

30

25

(d)

20

Local_SAR_5mm Average_SAR_5mm Local_SAR_10mm Average_SAR_10mm Local_SAR_15mm Average_SAR_15mm Local_SAR_20mm Average_SAR_20mm

15

10

0 0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

0,0

0,2

0,4

0,6

0,8

1,0

Norm alized distance

8
1,8 1,6 1,4 1,2

Normalized distance

(e)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

7 6

(f)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

SAR (W/Kg)

1,0 0,8 0,6

SAR (W/Kg)

5 4 3 2

0,4 0,2 0,0 0,0 0,2 0,4 0,6 0,8 1,0

1 0 0,0 0,2 0,4 0,6 0,8 1,0

Normalized distance

Normalized distance

596

Seddik Bri, Samira Kassimi, Mohamed Habibi and Ahmed Mamouni

Figure 6: SAR distribution in phantom, at different antenna-head distances, as layered sphere Skin, Bone and Muscle:
(a) Dipole at 900 MHz (c) Monopole at 900 MHz
20

(b) Dipole at 1800 MHz (d) Monopole at 1800 MHz

50

(a)
15

Local_SAR_5mm Average_SAR_5mm Local_SAR_15mm Average_SAR_15mm Local_SAR_25mm Average_SAR_25mm SAR in W/Kg

40

(b)

SAR in W/Kg

30

Local_SAR_5mm Average_SAR_5mm Local_SAR_15mm Average_SAR_15mm Local_SAR_25mm Average_SAR_25mm

10

20

5 10

0 0,0 0,2 0,4 0,6 0,8 1,0

0 0,0 0,2 0,4 0,6 0,8 1,0

Normalized distance
20 18 16 14

Normalized distance
60

(c)

SAR in W/Kg

SAR in W/Kg

12 10 8 6 4 2 0 0,0 0,2 0,4

Local_SAR_5mm Average_SAR_5mm Local_SAR_15mm Average_SAR_15mm Local_SAR_25mm Average_SAR_25mm

50

(d)

40

Local_SAR_5mm Average_SAR_5mm Local_SAR_15mm Average_SAR_15mm Local_SAR_25mm Average_SAR_25mm

30

20

10

0
0,6 0,8 1,0

0,0

0,2

0,4

0,6

0,8

1,0

Normalized distance

Normalized distance

Figure 7: Local and average SAR variation, at different antenna-head distances, for heterogeneous sphere Skin, Bone, Muscle and Brain:
(a) Dipole at 900 MHz (c) Monopole at 900 MHz (e) Patch at 900 MHz
20

(b) Dipole at 1800 MHz (d) Monopole at 1800 MHz (f) Patch at 1800 MHz
60

15

(a)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

50

(b)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

SAR in W/Kg

40

10

SAR in W/Kg

30

20

5
10

0 0,0 0,2 0,4 0,6 0,8 1,0

0 0,0 0,2 0,4 0,6 0,8 1,0

Normalized distance

Normalized distance

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies

597

12

10

(c)

SAR in W/Kg

SAR in W/Kg

Local_SAR_5mm Average_SAR_5mm Local_SAR_10mm Average_SAR_10mm Local_SAR_15mm Average_SAR_15mm Local_SAR_20mm Average_SAR_20mm

(d)
40

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

20

0 0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

0,0

0,2

0,4

0,6

0,8

1,0

Norm alized distance

4,0 2,0 1,8 1,6 1,4 2,5

Normalized distance

(e)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm SAR (W/Kg)

3,5 3,0

(f)

Local_SAR:5mm Average_SAR:5mm Local_SAR:20mm Average_SAR:20mm

SAR (W/Kg)

1,2 1,0 0,8 0,6 0,4 0,2 0,0 0,0

2,0 1,5 1,0 0,5 0,0 0,0

0,2

0,4

0,6

0,8

1,0

0,2

0,4

0,6

0,8

1,0

Normalized distance

Normalized distance

5. Interpretation
The value of the SAR is very important when the antenna is too near to the head and stripped when the antenna moves away from the head, for the two types of antennas. But the values of the SAR is very important for dipole antenna that for monopole. We deduce that the value of the SAR increases by 2,9 when we use a dipole antenna:

SARdipole = 2,9 SARmonopole

The SAR increases de 3,1 when the frequency is doubled: SAR900 = 3,1 SAR1800

(3)

(4) For simulation of several layers (skin, muscle, bone and brain), the field propagates in all the layers and the values of the SAR are strongly reduced in the bone, on the other side the great values are in the skin and the muscle. We also notice that the value of the SAR reduce when the antenna moves away from the head and when by using a monopole. The report between the levels of the SAR, when the antenna moves by 5mm, presented by the following formula: (5) With d is the distance between the antenna and the head. For the patch antenna, we find the values of SAR too small compared to other antenna (dipole and monopole), these values are close to the internationals standards. Other results of modelling by the

SAR ( d ) = 1,5 SAR ( d + 5)

598

Seddik Bri, Samira Kassimi, Mohamed Habibi and Ahmed Mamouni

software HFSS are presented in the table below, for different frequencies. The cellular telephone is modelled by a monopole or dipole and operating with various frequencies going from 300 MHz to 3000 MHZ and with a powerful 2 watt (table 2 above).
Table 2:
SAR values for 1g and 10g of tissue, and local SAR at excitation source y=0 and in y=2 cm.
1 g SAR 3,06 4,98 9,62 10,98 29,83 39,36 40,97 55,42 65,81 10 g SAR 2,07 3,31 6,62 7,02 16,50 20,45 21,21 25,42 26,48 Local SAR at Excitation source (y=0) 4,63 7,59 14,71 17,01 53,90 74,39 78,02 115 157.66 Local SAR at (y=2 cm) 2,10 3,28 4,93 5,47 6,54 6,85 6,54 8,09 8,8

F (MHZ) 300 450 835 900 1450 1800 1900 2450 3000

Notice that the 1g SAR is important than the 10 g SAR and the SAR increases as the frequency increases. The local SAR is important when the excitation source is close (y=0) and local SAR decrease moves away (y=2 cm). In order to properly compare the measured, the FDTD computed SAR distributions in the bowl, and the HFSS simulation, the FDTD values had to be calculated by averaging over several computational cells and E-field components All SAR values were normalized to 1W of radiated power [14][15]. The FDTD computed local SAR on the axis of symmetry in the spherical bowl when the antenna was placed symmetrically below it is shown in figure 8. The agreement between measurement and FDTD computation is very good for all distances between the bowl and the antenna. The peak local SAR is located at the inner surface of the bowl and falls off quite rapidly with increasing heightdistance from the inner surface.
Figure 8: FDTD computed SAR on the axis of symmetry for the spherical bowl at f=900 MHz. (Dipole was placed as centred 5, 25 and 50 mm below the outer South Pole)

The SAR decrease due to increased separation between the bowl and the dipole antenna is also clearly understandable and an increase in h from 5 mm to 50 mm decreases the maximum SAR almost by a factor of 10 both in the measurements and in the FDTD computations. The mean difference between the measurement and the FDTD data is 0,2 W/kg for h=5 mm, 0,1 W/kg for h=25 mm and

Specific Absorption Rate (SAR) Distribution in the Human Head at Global System Mobil (GSM) Frequencies

599

only 0,03W/kg for h=50 mm. The variation SAR with HFSS simulation is the same as the FDTD computation with a few differences.

6. Conclusion
Numerical modelling of human head-cellular phone is a very important problem. Analyzing possible range of variations of the induced field strengths in various tissues requires an extensive effort, since local field strengths strongly depend on various parameters: operational frequency, antenna power, mutual positions of the between device and head human, design of the device, size and the shape of human head, distribution of tissues within the head and the electrical properties of the tissues. So, the points listed below were especially mentioned in this paper: The values of the local and average SAR by HFSS and FDTD decrease with increasing depth Accuracy of these numerical simulations strongly depend on the reliability of electrical parameters of the tissues. Therefore, parallel studies should be continued with groups who conduct measurement researches. Finally, long term measurements, simulations and statistical evaluations should be done before making any declarations related to public health and safety. The feature of this study will be interested to a heterogeneous phantom radiated different antenna. Further works are required for evaluation of the temperature rise in the human head for portable phones with an anatomical model of the human head exposed to the field radiated by the user antenna in wireless frequencies.

Acknowledgment
This work is realised in the laboratory of the Institute of Electronics, the Microelectronics and Nanotechnology, Lille- France and was supported by the cooperation project MA: 153 / 06 between Morocco and France.

References
[1] S. Bri, A. Mamouni, L. Zenkouar, L. Bellarbi, M. Habibi, M. ELkadiri, ,Control of the temperature-depth profile in biological tissues : Medical Applications, International A. Physical Journal, Vol. 42, No.2, pp. 29-35, Belgium, 2001. S. Bri, L. Zenkouar, A. Saadi, L. Bellarbi, M. Habibi, A. Mamouni, Measure of the Temperature-Depth Profile by an S Band Radiometric Receiver for Biomedical Applications, International Journal Annals Telecoms, 59, Vol. 3-4, pp. 467-484, France, Mars-Avril 2004. O. Kivekas, J. Ollikainen, T. Lehtiniemi, Vainikainen, Bandwidth SAR and efficiency of internal mobile phone antennas, IEEE transactions On Electromagnetic Compatibility, Vol. 46, No. , pp. 71- 86, February 2004. H. Kawai, K.Ito,Simple Evaluation method of estimating local average SAR, IEEE Transactions on Microwave Theory and Techniques, Vol. 52,No. 8, pp.2021-2029, August 2004. ICNIRP, Guidelines for limiting exposure to time-varying electric, magnetic, and electromagnetic fields (up to 300 GHz), Health Phys., Vol. 74, No. 4, pp. 494-522, April 1998. A.Hirata, T.Asano, O. Fujiwara, FDTD analysis of human body-core temperature elevation due to RF far-field energy prescribed in the INCNIRP guidlines, Physics In Medecine and Biology, Vol. 52, pp. 5013-5023, 2007.

[2]

[3]

[4]

[5]

[6]

600 [7]

Seddik Bri, Samira Kassimi, Mohamed Habibi and Ahmed Mamouni O. P Gandhi, G. Lazzi, C. M. Furse, Electromagnetic absorption in the human head and neck for mobile telephones at 835 and 1900 MHz ,IEEE Trans. Microwave Theory Tech., Vol. 44, pp. 1884-1897,Oct.1996. A. O. Rodrigus, J. J. Viana, Calculation of temperature rise induced by cellular phones in the human head,Journal of microwaves and Optonelectronics, Vol. 6, No.1, pp. 310-322, 2007. Y. Okano, K. Ito, I. Ida, M. Takahashi, The SAR evaluation method by a combination of thermographic experiments and biological tissue-equivalent phantoms, IEEE Trans. MTT, Vol.48, No.11, pp. 2094-2103, Nov. 2000. A.O. Rodrigues, L. Malta, J.J. Viana, L.O.C. Rodrigues, J.A. Ramirez, "A head model for the calculation of SAR and temperature rise induced by cellular phones," IEEE Trans. on Magn, vol. 44, No.6, pp.1446-1449, 2008. S. Khalatbari, D. Sardari, A. A.Mirzaee and H. A. Sadafi, Calculating SAR in Two Models of the Human Head Exposed to Mobile Phones Radiations at 900 and 1800 MHz, Progress In Electromagnetic Research Symposium' 2006, Cambridge, USA, Vol. 2, No.1, pp.104-109, March 26-29, 2006. Karl S. Kunz, Raymond J. Luebbers, "The Finite Difference Time Domain Method for Electromagnetics", CRC Press, 1993. P. Stavroulakis, Biological Effects of Electromagnetic Fields, Springer, 2002. K. Caputa, M. Okoniewski and Maria A. Stuchly, "An Algorithm for Computations of the Power Deposition in Human Tissue with the FDTD Method", Proceedings of the USNC/URSI National Radio Science Meeting 1998, Atlanta, Georgia, U.S.A, June 21- 26, pp. 197, 1998. J. Wang, O. Fujiwara, FDTD computations of temperature rise in the human head for portable telephones, IEEE Trans. MTT, Vol.47, No.8, pp. 1528-1534, 1999.

[8] [9]

[10]

[11]

[12] [13] [14]

[15]